Shredder: Learning Noise Distributions to Protect Inference Privacy
cseweb.ucsd.edu · 6,231 words · saved by 1 readers
N/A
Shredder: Learning Noise Distributions to Protect Inference Privacy Fatemehsadat Mireshghallah Mohammadkazem Taram Prakash Ramrakhyani∗ Ali Jalali† Dean Tullsen Hadi Esmaeilzadeh {fmireshg,mtaram}@eng.ucsd.edu,prakash.ramrakhyani@arm.com,ajjalali@amazon.com,{tullsen,hadi}@eng.ucsd.edu Alternative Computing Technologies (ACT) Lab University of California San Diego ∗ Arm, Inc. † Amazon.com, Inc. Abstract…
saved by
related reading
- Defeating Nondeterminism in LLM Inference - Thinking Machines Labthinkingmachines.ai
- Private Cloud Compute: A new frontier for AI privacy in the cloud - Apple Security Researchsecurity.apple.com
- Can a Model Be Differentially Private and Fair?pair.withgoogle.com
- Bayesian Neural Networkscs.toronto.edu
- 1902.06101arxiv.org
- Transformer Math 101 | EleutherAI Blogblog.eleuther.ai
- VaultGemma: The world's most capable differentially private LLMresearch.google
- Nitrobrew: Fast, Lossless Distillation for Free | Tildeblog.tilderesearch.com
- Mind the Trust Gap: Fast, Private Local-to-Cloud LLM Chat · Hazy Researchhazyresearch.stanford.edu
- What is federated learning?research.ibm.com
- Differential Privacy: Issues for Policymakerssimons.berkeley.edu
- Combining Machine Learning and Homomorphic Encryption in the Apple Ecosystem - Apple Machine Learning Researchmachinelearning.apple.com